DENROCHE.

Charles
Text metaphtonymy - The interplay of metonymy and metaphor in discourse. Metaphor
and the Social World 8.1 (2018), pp. 1-24.

RESUMO:
Denroche (2018) mostra dos vários modos pelos quais o pensamento metafórico e o
metonímico, como fenômenos independentes, organizam o texto ao nível do discurso. A
literatura sobre metáfora no discurso classifica-a sob três categorias: “agrupamento”,
“cadeia” e “extensão”, o mesmo acontecendo com a metonímia. Denroche examina os
modos como os fenômenos da metáfora no discurso e da metonímia no discurso se
combinam para construir significado no nível do texto. A relação entre metáfora e
metonímia no discurso, referida aqui como metafotonímia textual, é explorada sob os
títulos adaptados de Goossens (1990), em especial, a “metáfora dentro da metonímia” e a
metonímia dentro da metáfora”. Esses modos de combinação no nível do discurso são
mostrados como sendo variados e intrincados. Esse fatos tem implicações para os
linguistas aplicados que trabalham com texto.

1 Introdução

Em 1990, Goossens cunhou o termo metafotonímia para referir-se à interaçao entre

metáfora e metonímia em expressões linguísticas. Goossens estava interessado não com
itens linguísticos que representam estágios intermediários no contínuo metáfora-
metonímia, mas com a metáfora e a metonímia como fenômenos distintos, aparecendo
“em combinação” e “entrelaçados”. Para ele, a expressão “metafotonímia textual” indica a
interação entre metáfora e metonímia não em unidades de comprimento de uma oração,
mas através de longos trechos da língua.
O objetivo do autor é a revisão dos diferentes odos pelos quais o pensamento figurativo
impacta o discurso no nível do texto.
Primeiramente, ele vê como a metáfora e a metonímia organizam o texto quando ocorrem
independentemente, e a seguir examina o mesmo fenômeno ocorrendo em combinação.
Ele oferece uma classificação do fenômeno tanto da metáfora-no-discurso quanto da
metonímia sob três categorias: “em grupo”, “em cadeia” e “em extensão”.
Ele oferece um enquadre que arranja os fenômenos em um número manipulável de
categorias. para mostrar os paralelos que existem entre as três categorias.

2. Metáfora no discurso
A literatura sobre metáfora no discurso envolve diferentes abordagens para entender o
papel da metáfora na construção do significado no nível do texto. Elas abrangem desde a
identificação sistemática e enumeração da metáfora linguística no texto, notando a
atividade local da metáfora em pontos críticos no texto (grupos), até a observação do
padronização de ligação de metáforas através do texto (cadeias) e, finalmente, a
metáforas sozinhas organizando longos trechos do texto ou trechos inteiros (extensão).
2.1 Agrupamento de metáforas
Os vários instrumentos criados por estudiosos no assunto ajudam na identificação de
agrupamentos de metáforas, concentrações de metáforas linguísticas ocorrendo em
proximidade significativa em certos pontos do texto. Essa distribuição da metáfora no
discurso é discutida por Darian (2000). Esses agrupamentos são reuniões de expressões
metafóricas do mesmo tema metafórico, desenvolvidos através de muitas sentenças ou
parágrafos. Para Cameron e Stelma (2004), eles envolvem metáforas convencionais ou
novas e podem derivar de um ou mais domínios fonte.
Geralmente, os agrupamentos referem-se a concentrações de metáforas linguísticas de
diferentes domínios, como define Semino (2008): “expressões metafóricas diferentes
oriundas de diferentes domínios fonte em alta proximidade entre si” (p.226). No
“agrupamento de metáforas misturadas” em texto jornalístico examinado por Kimmel
(2010), as metáforas aparecem em alta proximidade, mas apesar de serem de fontes não
relacionadas e portanto carente de coerência, elas não parece, apresentar problema de
processamento.
Os estudiosos concordam quanto à função dos agrupamentos> eles ocorrem onde um
trabalho discursivo intenso ou importante é realizado. Koller (2003) observa que os
agrupamentos no início de um texto tende a ter função ideacional, enquanto que no meio
do texto ou no final eles tendem a ser interpessoais (p.120). Para Deignan et al (2013)
agrupamentos são caracterizados como concentrações “mais elevadas do que a média”
em pontos do texto em que a mensagem é “particularmente difícil oi ameaçadores de
face” (pp8-9).
Em encontros de reconciliação (Cameron & Stelma, 2004), os agrupamentos oferecem
um meio de apresentar o “alteridade”; e na literatura sobre discurso e semões religiosos
sobre a relação entre psicoterapeuta e paciente explicam tóicos difíceis ou não familiares
(pp. 132-135).
2.2. Cadeia de metáfora

Uma “cadeia de metáfora” é um padrão de metáfora no discurso constituída por

metáforas relacionadas distribuídas de maneira mais ou menos uniformemente através do
texto. Koller (2003) e Semino (2008) usam o termo “cadeia” para descrever esse tipo de
padronagem, Para Semino (2008), as cadeias de metáforas são feitas de (geralmente
convencional) metáforas linguísticas oriundas de um único domínio fonte, “várias
expressões metafóricas relacionadas através do texto” e resultam de uma combinação de
“repetição”, “recorrência” e “extensão”. Koller (2003) identifica cadeias derivadas de
diferentes domínios, GUERRA, ESPORTE e JOGOS, em textos de marketing que ela
analisa, e mostra como as cadeias podem sobrepor-se e interagir sem necessariamente
criar problemas para o leitor. Em outro trabalho, Dariam (2000) usa o termo “metáfora
recorrente” para referir-se à “recorrência d mesma imagem em diferentes lugares do texto
(p.171), como padrões derivadas de IMMUNE SYSTEM AS WAR, GENETIC TRANSFER
AS FAMILY RELATIONS and BACTERIA AS HUNTERS in introductory science texts he
considers (Darian, 2000, PpoI71-172).
No exemplo abaixo, um trecho de artigo de jornal sobre a erformance relativa de duas
ocorrências, a libra e o euro, metáforas de orienetaçã/especial e metáforas de movimento,
derivadas principalmente de BAD IS DOWN, exerce um papel importante no
enquadramento da mensagem. Termos termos fonte relacionados ao domínio fonte
DOWN estão sublinhadas a seguir.

The pound's relentless slide towards parity with the euro picked up pace after it
plunged to another record low against the single European currency. The latest slide
saw sterling worth just 1.022euros amid expectations for European interest rates to
remain higher than in the UK […]. Sterling has lost 13% of its value against the euro
this month alone as it sinks to yet more historic lows […].
'Metro,' 29 December 2008 - http://metroocoouk/2008/12/29/pound-hits-nearparity-with-
euro-270090/)

2.3 Metáfora extendida

O terceiro fenômeno considerado sobre a metáfora no discurso é o que o autor

chama de “metáfora estendida”, que consiste em uma extensão de uma única metáfora
através de substancial porção do texto, ou mesmo do texto todo. Um exemplo em escala
pequena ocorrer em:

We have seen cuts in the health service not improvements, cuts that have
not only gone through the skin but have cut into the flesh and as far as the
bone in some cases. (Tcday' BBC Radio 4, 2012, author's transcription)

Darian (2000) caracterizaa a “metáfora estendida” como um ou vários parágrafos

que embelezam por meio de uma metáfora original e o transporta através de permutações”
como DNA IS A LIBRARY (Darian, 2000, P.171). Para ele, a função dessas metáforas é
heurístico, ajudando o leitor a “entender” e “lembrar-se” (2000, pp.168-169).
Para Semino (2008), “extensão (de metáforas linguísticas)” é a ocorrência de várias
expressões metafóricas evocando o mesmo domínio fonte e descrevendo o mesmo
domínio alvo em proximidade um do outro no texto” (227), Ele discute as seguintes
metáforas: PAIN CONTROL IS A GATE, BERLUSCONI IS A DISEASE and HAVING A
SPECIAL-NEEDS CHILD IS BEING SENT TO A HOLIDAY DESTINATION YOU DIDN'T
CHOOSE (Semino et al., 2013).
Há uma sobreposição potencial entre metáfora estendida e cadeia de metáfora, já
que ambas envolvem uma ideia metafórica única através de longos trechos de língua.
Goatly and Semino both see metaphor chains as manifestations of metaphor extension, and
link extension with organizing/systematic metaphor. For Goatly, extension involves different
vehicle terms from one domain (Goatly, 1997, P.264), which, when numerous, form
organizing metaphors (systematic metaphors), such as ANTS ARE SOLDIERS, and
contribute to 'textual structuring' (Goatly, 1997, P·163).
O exemplo seguinte (sobre os Jogos Olímpicos de 2012, que apareceu no London
Underground) é sem dúvida um exemplo de metáfora estendida. Há uma única ideia
metafórica que organia o texto todo. A metáfora envolvida é LONDON IS A FLAT. In the
'mark-up' below, language relating to the target domain LONDON is shown in bold and
language relating to the source domain FLAT is underlined:
 You know when your mum's coming round to your fiat and you give the place a
quick tidy? Well that's exactly what we're doing. Except our "fiat" is London and
our "mum" is the rest of the world coming round. So we're cleaning London
in time for the London 2012 Olympic Games. But that's a big job so we're
asking peopIe like you to Iend us a hand. We have litter to pick, graffiti to
scrub, and flowers to planto To help London look its best just go to
P&GCapitaIcleanup.com. Come on. Make your mum proud!
(Advertisement on the London Underground, [an 2012)

O texto começa com uma linguagem do domínio fonte, FLAT; há, então, uma
“transição” contendo linguagem tanto da fonte quando do alvo em que a metáfora é
explicada fazendo explícitos certos mapeamentos e até mesmo sinalizando-os usando dois
pontos: flat iguala-se a London, mãe iguala-se ao resto do mundo. Move-se, então, para a
linguagem do domínio alvo, LONDON; finalmente há um breve “retorno” para o domínio
fonte, Make your mum proud!
We have seen in this section that systematicity of metaphor use can produce
patterning of language in text of two types; it can result in metaphor chains if the language
and the metaphoric idea involved are conventional, or extended metaphor if the metaphoric
ideas involved are nove!. There will inevitably be contexts where the metaphoric idea is
somewhere between the two, producing patterns which are neither clearly chains nor
extensions.

This article starts by looking at the various ways metonymic and
metaphoric thinking, as independent phenomena, organize text at
discourse leve!.The literature on metaphor in discourse is
classified under three broad categories, 'metaphor clusters,
'metaphor chains' and 'extended metaphor, while the less
extensive body of research on metonymy in discourse is analyzed
into parallel categories, 'rnetonymy clusters, 'metonymy chains'
and 'extended metonymy'. The article goes on to look at the ways
in which metonymy-in-discourse and rnetaphor-in-discourse
phenomena combine in making meaning at text leve!.The
interplay of metonymy and metaphor in discourse, referred to here
as 'text metaphtonymy, is explored under headings adapted from
Goossens (1990), namely, 'metaphor within metonyrny' and
'metonymy within metaphor' The ways in which metonymy and
metaphor combine at discourse leveI are shown to be varied and
intricate. This has implications for applied linguists working with
text. The direction further work in this area might take is indicated.
Keywords: chains, clusters, extended metaphor, extended
metonymy, discourse, metaphor, metaphtonymy, metonymy, text;
Denroche (2018) mostra dos vários modos pelos quais o
pensamento metafórico e o metonímico, como fenômenos
independentes, organizam o texto ao nível do discurso. A
literatura sobre metáfora no discurso classifica-a sob três
categorias: “agrupamento”, “cadeia” e “extensão”, o mesmo
acontecendo com a metonímia.
Denroche examina os modos como os fenômenos da metáfora
no discurso e da metonímia no discurso se combinam para
construir significado no nível do texto. A relação entre metáfora
e metonímia no discurso, referida aqui como metafotonímia
textual, é explorada sob os títulos adaptados de Goossens
(1990), em especial, a “metáfora dentro da metonímia” e a
metonímia dentro da metáfora”. Esses modos de combinação no
nível do discurso são mostrados como sendo variados e
intrincados. Esse fatos tem implicações para os linguistas
aplicados que trabalham com texto.

***

1. Introduction 1 Introdução

In his 1990 article, Goossens coins the term metaphtonymy to
refer to the interaction of metonymy and metaphor in linguistic
expressions (Goossens, 1990).In that article, Goossens is
concerned not with linguistic items which represent intermediate
stages on the metonymy-metaphor continuum but with metonymy
and metaphor as distinct phenomena, appearing 'in cornbinatíon'
and 'intertwined' (Goossens, 1990, P.323). For him,
metaphtonymy is a phenomenon occurring on the smaJl scale of
individual expressions, and not metonymyand metaphor-led
phenomena on the larger scale of the whole text, which is the
focus of this article. In the present study, 1 will be using the term
'text metaphtonymy' to underscore this difference and indicate
Em 1990, Goossens cunhou o termo metafotonímia para referir-
se à interaçao entre metáfora e metonímia em expressões
linguísticas. Goossens estava interessado não com itens
linguísticos que representam estágios intermediários no
contínuo metáfora-metonímia, mas com a metáfora e a
metonímia como fenômenos distintos, aparecendo “em
combinação” e “entrelaçados”. Para ele, a expressão
“metafotonímia textual” indica a interação entre metáfora e
metonímia não em unidades de comprimento de uma oração,
mas através de longos trechos da língua.
that the focus is the interaction of metonymy and metaphor not
within clause-length units but across longer stretches of language.
The purpose of this article is to review the different ways
in which figurative thought impacts on discourse at text level, To
do so, 1 demonstrate the various ways in which figurative thought
manifests itself in speech/writing. 1 look first at how metaphor and
metonymy organize talk!text when occurring independently, and
then look at the same phenomena occurring in combination.
Section 2 offers a classification of metaphor-in-discourse
phenomena under three broad categories - 'metaphor clusters,
'metaphor chains' and 'extended metaphor, while Section 3 deals
with the less-studied topic of metonymy-in-discourse under three
parallel categories - 'metonymy clusters, 'metonymy chains' and
'extended metonymy' Every one of these six categories is
represented in some form in the literature, though often named
differently. What 1 offer is a framework which overviews/arranges
the phenomena into a manageable number of categoríes, named
to show up the parallels which exist between the three metaphor
phenornena and the three metonymy phenomena. This involves
a fresh look at terrninology, but not merely as an exercise in re-
naming, the framework is not an end in itself but, rather, a tool of
investigation of the phenomenon at the centre of this study, 'text
metaphtonymy' Section 4 examines text metaphtonymy, the co-
occurrence of metaphor and metonymy in talk/text. Section 5
reviews the contributionthe article makes to the field and suggests
the direction further research in this area might take.
***
2. Metaphor in discourse
The sizeable literature on metaphor in discourse
encompasses a variety of different approaches to understanding
the role of metaphor in meaning-making at text leveI. They range
from the systematic identification and enumeration of linguistic
metaphor in text, to noticing local metaphor activity at critical
points in texts (clusters'), to observing the patterning of metaphors
linking across a text (chains') and, finally, to single metaphors
organizing long stretches of text and whole texts ('extended
metaphor').
2.1 Metaphor clusters
A number of invaluable tools have been developed for
identifying metaphor in discourse, such as those devised by
Cameron (2003), Cameron & Deignan (2006), Cameron & Maslen
(2010), Pragglejaz Group (2007), Steen (2002, 2007) and Steen
et al. (2010). The Metaphor Identification Procedure (M1P)
operationalizes metaphor identification at word level by identifying
words used metaphorically which have a more basic meaning
(Pragglejaz Group, 2007). A modified version of M1P, M1PVU,
includes similes, comparisons and extended comparisons (Steen
et al., 2010). In Steens five-step procedure for metaphor
identification, the clause is the unit of analysis (Steen, 2002,
2007). Metaphor Identification through Vehicle (M1V) analyses
discourse for both single and multi-word 'vehicle terms' (Cameron,
1999, 2003), using the full intonation unit as the unit of analysis
(Cameron & Stelma, 2004, p.riç: Cameron et al., 2010). The focus
on 'ernergent' meaning is central to the 'discourse-dynamics
approach' to metaphor analysis (Cameron et al., 2009; Cameron
& Maslen, 2010). It enables researchers to recognize subtle,
locally-occurring and often ephemeral metaphor activity, as well
as more stable 'metaphoremes, units which show shared features
of forrn, sernantics, affect and pragmatics (Cameron & Deignan,
2006, p. 676).
These identification tools have been helpful in the
identification of metaphor 'clusters, concentrations of linguistic
metaphors occurring in close proximity at particular points in a
text. This uneven distribution of metaphor in discourse is
discussed by Darian (2000), Koller (2003), Cameron & Low
(2004), Cameron & Stelma (2004), Cameron (2008), Semino
(2008) and Kimmel (2010). For Darian, 'clusters' are groupings of
metaphoric expressions from the same metaphoric theme,
developed over several sentences or paragraphs (Darian, 2000,
pp. 180-181). For Cameron & Stelma, they involve conventional
or novellinguistic metaphors and can derive from one or a number
of 'vehicle' domains (Cameron & Stelma, 2004). More usually,
O objetivo do autor é a revisão dos diferentes odos pelos quais
o pensamento figurativo impacta o discurso no nível do texto.
Primeiramente, ele vê como a metáfora e a metonímia
organizam o texto quando ocorrem independentemente, e a
seguir examina o mesmo fenômeno ocorrendo em combinação.
Ele oferece uma classificação do fenômeno tanto da metáfora-
no-discurso quanto da metonímia sob três categorias: “em
grupo”, “em cadeia” e “em extensão”.
Ele oferece um enquadre que arranja os fenômenos em um
número manipulável de categorias. para mostrar os paralelos
que existem entre as três categorias.
2. Metáfora no discurso
A literatura sobre metáfora no discurso envolve diferentes
abordagens para entender o papel da metáfora na construção
do significado no nível do texto. Elas abrangem desde a
identificação sistemática e enumeração da metáfora linguística
no texto, notando a atividade local da metáfora em pontos
críticos no texto (grupos), até a observação do padronização de
ligação de metáforas através do texto (cadeias) e, finalmente, a
metáforas sozinhas organizando longos trechos do texto ou
trechos inteiros (extensão).
2.1 Agrupamento de metáforas
Os vários instrumentos criados por estudiosos no assunto
ajudam na identificação de agrupamentos de metáforas,
concentrações de metáforas linguísticas ocorrendo em
proximidade significativa em certos pontos do texto. Essa
distribuição da metáfora no discurso é discutida por Darian
(2000). Esses agrupamentos são reuniões de expressões
metafóricas do mesmo tema metafórico, desenvolvidos através
de muitas sentenças ou parágrafos. Para Cameron e Stelma
(2004), eles envolvem metáforas convencionais ou novas e
podem derivar de um ou mais domínios fonte.
Geralmente, os agrupamentos referem-se a concentrações de
metáforas linguísticas de diferentes domínios, como define
Semino (2008): “expressões metafóricas diferentes oriundas de
diferentes domínios fonte em alta proximidade entre si” (p.226).
No “agrupamento de metáforas misturadas” em texto jornalístico
examinado por Kimmel (2010), as metáforas aparecem em alta
proximidade, mas apesar de serem de fontes não relacionadas
e portanto carente de coerência, elas não parece, apresentar
problema de processamento.
Os estudiosos concordam quanto à função dos agrupamentos>
eles ocorrem onde um trabalho discursivo intenso ou importante
é realizado. Koller (2003) observa que os agrupamentos no
início de um texto tende a ter função ideacional, enquanto que
no meio do texto ou no final eles tendem a ser interpessoais
(p.120). Para Deignan et al (2013) agrupamentos são
caracterizados como concentrações “mais elevadas do que a
média” em pontos do texto em que a mensagem é
“particularmente difícil oi ameaçadores de face” (pp8-9).
'clusters' refer to concentrations of linguistic metaphors from
different domains, as in Sernino's definition: "different
metaphorical expressions drawing from different source domains
in close proximity to one another" (Sernino, 2008, p.226). In the
'mixed metaphor clusters' in newspaper texts which Kimmel
díscusses, metaphors appear in particularly close proximity, but
in spite ofbeing from unrelated sources and therefore lacking
obvious coherence, they do not seem to present processing
problems (Kimmel, 2010).
Scholars agree as to the function of clusters: they occur
where intense or important discourse work is being done. Koller
observes that clusters at the beginning of a text will tend to have
an ideational function, while the function of clusters mid-text or at
the end of a text will tend to be interpersonal (Koller,2003, p.120).
In the literature reviewed by Deignan et al., clusters are
characterized as 'higher than average' concentrations of
metaphor, occurring at points in text where the message is
"particularly difficult or face threatening" (Deignan et al., 2013, pp.
8-9). For Cameron, they "rnark points in talk where something
complex or unfamiliar needs to be explained or interpreted"
occurring on "both micro and Em encontros de reconciliação (Cameron & Stelma, 2004), os
macro scales of talk", that is, from three to four intonation units agrupamentos oferecem um meio de apresentar o “alteridade”; e
to passages lasting na literatura sobre discurso e semões religiosos sobre a relação
minutes (Cameron, 2008, po200)0 Cameron & Stelma observe entre psicoterapeuta e paciente explicam tóicos difíceis ou não
bursts occurring at critical junctures in communication, "points familiares (pp. 132-135).
where intensive and important discourse work is carried out"
(Cameron & Stelma, 2004, po13S)0In the reconciliation
encounters considered by Cameron & Stelma, clusters provide a
way of presenting 'otherness'; and in the literature on
psychotherapist-patient discourse and religious sermons they
review, of explaining difficult or unfamiliar topics (Cameron &
Stelma, 2004, PpoI32-

***

2.2 Metaphor chains 2.2. Cadeia de metáfora

A 'metaphor chain' is a metaphor-in-discourse pattern
made up of related metaphors distributed more or less evenly
across a text. Koller (2003) and Semino (2008) both use the
term 'chain' to describe this type of patterning. For Semino,
metaphor chains are made up of (usually conventional)
linguistic metaphors from a single source domain, "several
related metaphorical expressions throughout a text', and result
from a combination of 'repetition, 'recurrence' and 'extension'
(Semino, 2008, po226)0
Koller identifies chains deriving from different domains, WAR,
SPORTS and GAMES, in the marketing text she analyses
(Koller, 2003), and shows how chains can overlap and interact
without necessarily creating problems for the reader, In earlier
work, Darian uses the term 'recurring metaphor' to refer to the
"recurrence of the same image at different places in the text"
(Darian, 2000, po171), such as patterns deriving from IMMUNE
SYSTEM AS WAR, GENETIC TRANSFER AS FAMILY
RELATIONS and BACTERIA AS HUNTERS in introductory
science texts he considers (Darian, 2000, PpoI71-172)0
In the example I offer below, an extract from a
newspaper article on the relative performance of two currencies,
the pound and the euro, orientational/spatial metaphors and
metaphors of movement, deriving mainly from BAD IS DOWN,
play an important role in framing the message. Vehicle terms
relating to the source domain DOWN are shown underlined
below,
The pound's relentless slíde towards parity with
the euro picked up pace after it plunged to
another record kllir against the single European
currency. The latest slide saw sterling worth just
1.022euros amid expectations for European
interest rates to remain higher than in the UK
[oo0]0Sterling has lost 13%of its value against
the euro this month alone as ít sinks to yet more
historic lows […].
Ç'Metro,' 29 December 2008 - this month alone as it sinks to yet more historic lows […].
http://metroocoouk/2008/12/29/pound-hits-
nearparity-with-euro-zzcoço/)
Uma “cadeia de metáfora” é um padrão de metáfora no discurso
constituída por metáforas relacionadas distribuídas de maneira
mais ou menos uniformemente através do texto. Koller (2003) e
Semino (2008) usam o termo “cadeia” para descrever esse tipo de
padronagem, Para Semino (2008), as cadeias de metáforas são
feitas de (geralmente convencional) metáforas linguísticas
oriundas de um único domínio fonte, “várias expressões
metafóricas relacionadas através do texto” e resultam de uma
combinação de “repetição”, “recorrência” e “extensão”. Koller
(2003) identifica cadeias derivadas de diferentes domínios,
GUERRA, ESPORTE e JOGOS, em textos de marketing que ela
analisa, e mostra como as cadeias podem sobrepor-se e interagir
sem necessariamente criar problemas para o leitor. Em outro
trabalho, Dariam (2000) usa o termo “metáfora recorrente” para
referir-se à “recorrência d mesma imagem em diferentes lugares
do texto (p.171), como padrões derivadas de IMMUNE SYSTEM
AS WAR, GENETIC TRANSFER AS FAMILY RELATIONS and
BACTERIA AS HUNTERS in introductory science texts he
considers (Darian, 2000, PpoI71-172).
No exemplo abaixo, um trecho de artigo de jornal sobre a
erformance relativa de duas ocorrências, a libra e o euro,
metáforas de orienetaçã/especial e metáforas de movimento,
derivadas principalmente de BAD IS DOWN, exerce um papel
importante no enquadramento da mensagem. Termos termos
fonte relacionados ao domínio fonte DOWN estão sublinhadas a
seguir.
The pound's relentless slide towards parity with the euro
picked up pace after it plunged to another record low
against the single European currency. The latest slide
saw sterling worth just 1.022euros amid expectations for
European interest rates to remain higher than in the UK
[…]. Sterling has lost 13% of its value against the euro
'Metro,' 29 December 2008 -
http://metroocoouk/2008/12/29/pound-hits-nearparity-with-euro-
270090/)
Metaphoric mappings related to BAD IS DOWN have a
significant impact across a long stretch of this text; however, the
metaphoric senses of the words slide, plunged; low, higher,
sinks, laws are conventional, well established meanings in the
corpus of the language. The genre also limits choice, so that in
this type of news reporting these words are almost unavoidable.
In newspaper reports of the 2008 financial crisis we would no
doubt find many other words deriving from BAD IS DOWN, such
as collapse, slump, dive.fall, tumble, used not creatively but in
straight-forward reporting, as if, to quote Cameron et al., "the
metaphorical way of talking about it has become so
conventionalized that it is almost the only way to talk about it"
(Cameron et al., 2010, p.127). If the organizing metaphor is a
'primary conceptual metaphor, in other words, one dose to our
physical experience of the world, such as BAD IS DOWN, a
chain of lexical items of this sort typically results. If the
organizing metaphor is complex or novel, different lexical
patternings emerge for which the term 'chain' is no longe r
appropriate. These are more
likely to be examples of'extended metaphor' (Section 2.3).
Other terms which have been used to describe chains
include 'metaphor theme' (Musolff, 2000), 'metaphor formula'
(Kimmel, 2012) and 'recurrent metaphor' (Low, 2008), but the
most used and widely discussed term in this context is
'systematic metaphor' (Cameron, 2008; Cameron & Maslen,
2010). Cameron et aI. define systematic metaphor as "a set of
linguistic metaphors in which connected vehicle words or
phrases are used metaphorically about a particular topic"
(Cameron et al., 2010, p.127). "The systematic use of
connected metaphors across talk" forms a larger 'trajectory' or
'trace' (Cameron et al., 2009, p. 77), thereby constructing a
'metaphor trajectory' inside the 'discourse trajectory'
(Cameron, 2010, p. 84).
This 'discourse dynamics' perspective of Cameron
and her co- researchers is concerned with metaphor which is
"processual, emergent, and open to change" (Cameron et al.,
2009, p.67), where 'systematic metaphor' is "the dynamic
collection of connected linguistic metaphors, a trajectory from
one metaphor to the next over the dynamics of talk" (Cameron
et al., 2009, P.78). We are warned in this approach against
over-interpreting data and over-generalizing beyond the text
(Cameron et al., 2010, pp.1l9, 124-125, 138). Systematic
metaphors are less generalized than conceptual metaphors,
describing choices re!ating to specific texts and genres
(Cameron, 2008, p.208; 2010, p.129), coming closest to
conceptual metaphors only when "highly conventionalized
linguistic metaphors [... ] fali into highly conventionalized
patterns of use" (Cameron et al., 2010, P.134).
The terms 'conceptual metaphor' and 'systernatic
metaphor' reflect different priorities and different schools of
thought: conceptual metaphor theory (CMT) asks broad
questions about metaphor in language in the mind, while
metaphorled discourse analysis (MLDA) asks more specific
questions about metaphor in relation to context and the role of
mutual relationships, identities and the culture of the
participants in a specific speech event (Semino, 2008, P.31).
Semino distinguishes between 'discourse systematicity (of
metaphors)' and 'global systematicity (of metaphors), "the
conventional use of a set of related metaphorical expressions"
within a specific genre/discourse and across
genres/díscourses, respective!y (Semino, 2008, pp.227, 228).
Both generalize about metaphor, but while conceptual
metaphors record higher-Ievel generalizations about permanent
crossdomain mappings in the conceptual systems in our rninds,
systematic metaphor describes local use by language
participants while 'talking and thinking' in a specific discourse
event (Carneron, 2003; Cameron & Maslen, 2010). The
typographical
convention of writing conceptual metaphors in non-italic SMALL
CAPITALS and systematic metaphors in italic SMALL
CAPITALS (Cameron et al., 2010, P.lI7)
underscores this difference.

***
2.3 Extended metaphor 2.3 Metáfora extendida

The third phenomenon considered in this overview of metaphor in O terceiro fenômeno considerado sobre a metáfora no discurso
discourse, following the literature, I am calling 'extended é o que o autor chama de “metáfora estendida”, que consiste
rnetaphor' This is the nove! extension of a single metaphoric idea em uma extensão de uma única metáfora através de substancial
across a substantial portion of text, or even an entire text. An porção do texto, ou mesmo do texto todo. Um exemplo em
example of this occurring on a small scale is given below: escala pequena ocorrer em:
We have seen cuts in the health service not
improvements, cuts that have not only gone
through the skin but have cut into the flesh We have seen cuts in the health service not
and as far as the bone in some cases. improvements, cuts that have not only gone
(Tcday' BBC Radio 4, 2012, author's through the skin but have cut into the flesh
transcription) and as far as the bone in some cases.
(Tcday' BBC Radio 4, 2012, author's
Scholars who discuss extension include Goatly (1997), Darian transcription)
(2000), Steen (2007) and Semino (2008). Darian characterizes
'extended metaphor' as "one or several sequential paragraphs
that embellish on an original metaphor and carry it through several
perrnutations" such as DNA IS A LIBRARY (Darian, 2000, P.171). Darian (2000) caracterizaa a “metáfora estendida” como um ou
He sees the function of such metaphors as heuristic, helping the vários parágrafos que embelezam por meio de uma metáfora
reader 'understand' and 'remernber' (2000, pp.168-169). original e o transporta através de permutações” como DNA IS A
'Extension of rnetaphor' is the third of Steens 'four dimensions of LIBRARY (Darian, 2000, P.171). Para ele, a função dessas
metaphor in usage; the others being 'directness, 'signalling' and metáforas é heurístico, ajudando o leitor a “entender” e “lembrar-
'explicitness' (Steen, 2007, pp. 319-323). He observes that se” (2000, pp.168-169).
'metaphor extension' is processed differently from 'restricted
metaphor, where metaphor is confined to a discourse unit, in
terms of cross-domain mapping (Steen, 2007, p. 321).
For Semino, 'extension (oflinguistic rnetaphors)' is the
occurrence of"several
metaphorical expressions evoking the same source domain and
describing the same target domain in dose proximity to one
another in a text" (Sernino, 2008, p.227). The size of unit extends
to whole texts and to groups of related texts in Semino et al.s
discussion of texts drawing on PAIN CONTROL IS A GATE, Para Semino (2008), “extensão (de metáforas linguísticas)” é a
BERLUSCONI IS A D1SEASE and HAVING A SPECIAL-NEEDS ocorrência de várias expressões metafóricas evocando o
CHILD IS BEING SENT TO A HOLIDAY DESTINATION YOU mesmo domínio fonte e descrevendo o mesmo domínio alvo em
DIDN'T CHOOSE (Semino et al., 2013). They show how the proximidade um do outro no texto” (227), Ele discute as
metaphors used to frame the original texts offer possibilities for seguintes metáforas: PAIN CONTROL IS A GATE,
subsequent 'recontextualization' when contributors develop ('re- BERLUSCONI IS A DISEASE and HAVING A SPECIAL-NEEDS
frame') the original metaphor creatively through 'use and reuse' in CHILD IS BEING SENT TO A HOLIDAY DESTINATION YOU
blogs and in online fora (Semino et al., 2013, PP.46-51). Deignan DIDN'T CHOOSE (Semino et al., 2013).
et al., considering metaphor extending across a range of genres
and registers, such as climate change, and children and staff in a
nursery context, similarly show how GENE REPLICATION IS
COPYING and CONTROL OF PAIN IS A GATE give rise to
differently nuanced meanings when taken up in specialized or
popular genres (Deignan et aI., 2013).
There is a potential overlap between extended
metaphor and metaphor chains, as they both involve a single
metaphoric idea over a long stretch of language. Goatly and
Semino both see metaphor chains as manifestations of metaphor
extension, and link extension with organizing/systematic
metaphor. For Goatly, extension involves different vehicle terms
from one domain (Goatly, 1997, P.264), which, when numerous,
form organizing metaphors (systematic metaphors), such as
ANTS ARE SOLDIERS, and contribute to 'textual structuring'
(Goatly, 1997, P·163).
The example below, a poster which appeared on the
London Underground to recruit volunteers for the London 2012
Olympic Games, is unambiguously an example of extended Há uma sobreposição potencial entre metáfora estendida e
metaphor rather than a chain. It is a particular kind of extended cadeia de metáfora, já que ambas envolvem uma ideia
metaphor, though, where a single novel metaphoric idea metafórica única através de longos trechos de língua. Goatly
organizes the whole text, structuring it into a number of clearly and Semino both see metaphor chains as manifestations of
defined stages. The metaphor involved is LONDON IS A FLAT. In metaphor extension, and link extension with
the 'mark-up' below, language relating to the target domain organizing/systematic metaphor. For Goatly, extension involves
LONDON is shown in bold and language relating to the source different vehicle terms from one domain (Goatly, 1997, P.264),
domain FLAT is underlined: which, when numerous, form organizing metaphors (systematic
You know when your mum's coming round to your metaphors), such as ANTS ARE SOLDIERS, and contribute to
fiat and you give the place a quick tidy? Well that's 'textual structuring' (Goatly, 1997, P·163).
exactly what we're doing. Except our "fiat" is O exemplo seguinte (sobre os Jogos Olímpicos de 2012, que
London and apareceu no London Underground) é sem dúvida um exemplo de
our "mum" is the rest of the world coming round. metáfora estendida. Há uma única ideia metafórica que organia o
So we're cleaning London in time for the London texto todo. A metáfora envolvida é LONDON IS A FLAT. In the
2012 Olympic Games. But that's a big job so 'mark-up' below, language relating to the target domain LONDON
we're asking peopIe like you to Iend us a hand. is shown in bold and language relating to the source domain FLAT
We have litter to pick, graffiti to scrub, and flowers is underlined:
to planto To help London look its best just go to
P&GCapitaIcleanup. com. Come on. Make your
mum proud!
(Advertisement on the London Underground, [an
2012)
You know when your mum's coming round to your
fiat and you give the place a quick tidy? Well that's
 The text starts with language from the source domain, FLAT; there exactly what we're doing. Except our "fiat" is
is then a 'transition' containing language from both the source and London and
target in which the metaphor is explained by making certain our "mum" is the rest of the world coming round.
mappings explicit, even signalling them using quotation marks: So we're cleaning London in time for the
flat equals London, mum equals the rest of the world. It then London 2012 Olympic Games. But that's a big
moves on to language from the target domain, LONDON; and job so we're asking peopIe like you to Iend us
finally there is a brief 'return' to the source domain, Make your a hand. We have litter to pick, graffiti to scrub,
mum proud! We sense here that this is deliberate metaphor use, and flowers to planto To help London look its
one which is 'worked at' consciously in a way rarely achieved in best just go to P&GCapitaIcleanup.com.
speech, with the result that metaphor does not just pattern lexis, Come on. Make your mum proud!
but constructs a sequence of clearly identifiable moves: (Advertisement on the London Underground, [an
SOURCETRANSITION-TARGET-RETURN. The terms 'extended 2012)
metaphor' and systematic metaphor' hardly seem adequate to
describe this kind of metaphor-in-discourse phenomenon; a term O texto começa com uma linguagem do domínio fonte, FLAT;
such as 'text-constructing metaphor' or 'genre-constructing há, então, uma “transição” contendo linguagem tanto da fonte
metaphor' might be more appropriate. quando do alvo em que a metáfora é explicada fazendo
Paradoxically, the larger the unit of language organized explícitos certos mapeamentos e até mesmo sinalizando-os
by metaphoric thought, and, therefore, in a sense, the more usando dois pontos: flat iguala-se a London, mãe iguala-se ao
important the role of metaphor, the less likely it is that the resto do mundo. Move-se, então, para a linguagem do domínio
language will be identified as metaphoric, using tools such as MIP, alvo, LONDON; finalmente há um breve “retorno” para o domínio
MIPVU and MIV, or tagging software for automated analysis (e.g., fonte, Make your mum proud!
Deignan, 200sa, 200Sb; Kimmel, 2012;Stefanowistch & Gries,
2006). These procedures are well suited to the identification of
phenomena where metaphor is expressed as linguistic metaphor, We have seen in this section that systematicity of metaphor use
such as metaphor clusters, metaphor chains and emergent can produce patterning of language in text of two types; it can
metaphor, but not so well equipped for identifying metaphoric result in metaphor chains if the language and the metaphoric
thought operating on larger units. When metaphor organizes idea involved are conventional, or extended metaphor if the
substantial stretches oflanguage, linguistic metaphors may not metaphoric ideas involved are nove!. There will inevitably be
actually be present, as metaphoric writing does not necessarily contexts where the metaphoric idea is somewhere between the
contain local metaphor. Because metaphor-identification two, producing patterns which are neither clearly chains nor
procedures work at the small-scale end of analysis, when applied extensions.
to a text such as the Olympics Games text discussed above, the
overarching metaphor which constructs this text would not be
detected; in fact, only two words would be identified as
metaphoric, flat and mum. But if the text is marked up for source
and target language (in the way shown above), the entire text
becomes highlighted.
Metaphor-identification protocols detect metaphor
across discourse by identifying individual metaphoric
expressions, but metaphor across text is present in other
ways.This is not to say that scholars working on the identification
of linguistic metaphor are unaware of larger-scale phenomena.
On the contrary, Cameron recognises systematicity at three
levels: local, discourse and global (Cameron, 1999);while Steen
recognizes word, utterance, text and discourse levels of analysis
(Steen, 2014). Steen describes text patterns found in education,
science, advertising and propaganda, as well as literature, with
two clearly defined sections, where "some cross-domain
mappings are expressed as a text or section of a text with two
different parts, one of which is devoted to the source domain and
the other to the target domain" (Steen, 2007, P.342). He notes
also that "extended comparison typically has relatively long
stretches of direct language use for one domain followed by long
stretches of direct language use for another dornain" (Steen,
2007, p.321). It is this sort of 'direct language' which identification
procedures are not well equipped at detecting.
We have seen in this section that systematicity of
metaphor use can produce patterning of language in text of two
types; it can result in metaphor chains if the language and the
metaphoric idea involved are conventional, or extended metaphor
if the metaphoric ideas involved are nove!. There will inevitably be
contexts where the metaphoric idea is somewhere between the
two, producing patterns which are neither clearly chains nor
extensions.

.
3. Metonymy in discourse 3. Metonímia no discurso
The literature on metonymy in discourse, which I overview in this
section, is far less extensive than that on metaphor in discourse.
This reflects the greater interest in metaphor in studies of
figurative language/thought in general and that, historically, it
was metaphor which led the way in driving the 'cognitive turn'
Typically, metonymy occupies one chapter in books otherwise
devoted to metaphor, such as Lakoff & Johnson (1980), Gibbs
(1994), and Kõvecses (2002). The multiauthored volumes of
collected essays which form the backbone of the 'metonyrny
studies' literature, such as Benczes et al. (20n), Panther &
Radden (1999) and Panther & Thornburg (2003), are rich in their
discussions of clause-level phenornena, but give little attention
to metonymy at discourse levei; and collections with 'rnetonymy'
and 'metaphor' in the title, such as Barcelona (2000), Dirven &
Põrings (2003) and Panther et al. (2009), while redressing the
balance by giving plenty of room to discussions of metonymy,
give little space to how metonymy and metaphor interact at
discourse level,

The literature relevant to the present study

is, nonetheless, far from sparse. In
overviewing metonymy in discourse, in
addition to writings by cognitive linguists, I
consider Al-Sharafis (2004)
multidisciplinary, text-linguistics approach,
as well as work from semiotics (Iakobson,
1956) and literary linguistics (Lodge,
1977). The headings I adopt mirror the
categories in Section 2.

***

3.1 Metonymy clusters 3.1 Metonymy clusters

The term 'metonymy cluster' usually refers

in the metonymy literature to points in
discourse where linguistic metonymies are
found in high density, metonymies of the
kind which can be identified using
metonymy identification procedures such
as those developed by Biernacka (2013)
and Deignan et al. (2013). In this section, I
am using 'metonymy cluster' in a different
sense, referring instead to a group of
carefully chosen examples. I am calling
these 'metonyrnies' because they are
individual, specific, usually prototypical,
instances which convey a more general
message. Thus metonymic reasoning is
involved but on a larger scale, with the
result that the clusters I am identifying
would not necessarily involve metonymic
language and would therefore not be
identifiable using the procedures cited
above.
A 'cluster' in my sense usually takes Denroche (2018) usa o termo ‘cluster” no
the form of a list; but it is a metonymic list sentido de uma lista; mas é uma lista
rather than a 'literal' list, such as a shopping metonímica e não uma lista “literal”.Por
list or an inventory. For example, if a text exemplo, se um texto contem uma lista
contained an exhaustive list of ali the exaustiva de todas as faciclidades de um
facilities a gym or hotel had to offer, or ali ginásio ou de um hotel ou as coisas que
the things that take place in a village as part
of seasonal festivities, these would be
literallists (checklists), not metonymic lists,
and we would expect to process them
literally. If, however, a text advertising what
there is to do in a shopping mall read as
follows, You can buy a new evening dress,
have a teppanyaki meal with friends ar
attend the premiere of a HoUywood [ilm,
this is a metonymic list, a metonymy cluster,
as the three examples are not ali the
possibilities open to a visitor to the mall, and
require the reader to process them
metonymically. The impact of this cluster is
different from a generic phrase, such as The
mall offers retail, dining and entertainment
possibilities. Clusters of this sort are a
common and powerful rhetorical device.
The effect is to reinforce the argument by
bringing the hearerlreader into closer
physical proximity with the situation being
evoked through a register which is more
vivid and real.
The extract below is from a radio
interview with a British bishop about the
increased use offoodbanks (centres for
distributing food to the needy) in the UK. It
starts with a metonymy cluster consisting of
two sentences (underlined):
Bishop Walker: What we're
finding is that this is about older
people who are forced to
choose between having the
heating on or having breakfast.
It's about children whose mums
are faced with deciding who's
going to go without a meal that
day. Being on the breadline
used to be a bit of a political
metaphor, For half a million
Britons it's now a tragic truth
and the report we'vejust heard
simply bears that out. ('The
World at One, BBC Radio 4, 20
February 2014)
The cluster gives two instances of people in
society finding it hard to cope, older people
and mothers. Again it demands of the
hearerlreader that the language involved is
processed metonymically, as non-literal: if
understood literally, the issue discussed
acontecem em uma vila como parte de
festividades de estação, esta seriam lista
de checagem ou literallist, não uma lista
metonímica, e nos a processaríamos
literalmente. Se, contudo, um texto
avisando o que há para fazer em um
shopping seria: You can buy a new
evening dress, have a teppanyaki meal
with friends ar attend the premiere of a
HoUywood [ilm, esta é uma lista
metonímica, um cluster metonímico, já que
os três exemplos não são todas as
possibilidades abertas para o visitante, e
requer que o leitor processe
metonimicamente. O impacto desse
cluster é diferente de uma frase genéric,
como The mall offers retail, dining and
entertainment possibilities. Cluster deste
tipo são comuns e instrumento retórico
poderoso. O efeito é reforçar o argumento
trazendo o leitor para uma proximidade
física com a situação sendo evocada por
meio de registro que é mais vívido e real.
O extrato abaixo vem de uma entrevista
radiofônica com um bispo inglês sobre o
uso crescente de centros de distribuição
de comida para os necessitados). Ele
começa com um cluster metoníico
consistindo de duas sentenças
Bishop Walker: What we're
finding is that this is about older
people who are forced to
choose between having the
heating on or having breakfast.
It's about children whose mums
are faced with deciding who's
going to go without a meal that
day. Being on the breadline
used to be a bit of a political
metaphor, For half a million
Britons it's now a tragic truth
and the report we'vejust heard
simply bears that out. ('The
World at One, BBC Radio 4, 20
February 2014)
O cluster dá dois exemplos de pessoas na
sociedade finding it hard to cope, velhos e
mães. Novamente exige do leitor que a
língua envolvida é processada
metonimicamente, como não-literal: se
entendida literalmente, a questão
would seem to concern a much narrower discutida pareceria referir-se a tópico
topic, just two specific contexts. There are muito mais estreito, apenas dois contextos
fewer components here than in the earlier específicos. Há poucos componentes aqui
shopping mall example, two rather than do que no exemplo sobre o shopping, dois
three, and they are longer. A cluster may e não três, e são longos. A cluster pode
consist of a single instance and one which consistir de uma único exemplo e outro
may be quite extensive. que pode ser bem extensivo.
It is not only the number but also the Não é apenas o número, mas
type of items which acts as a trigger; items também o tio de item que atua como
in a metonymic list are prototypical, as in trigger;
this extract from a newspaper article:
Compare 2000 London with the
thin flame of Sixties Swinging
London: then,
there were only The Beatles,
Carnaby Street, King's Road,
Australians in damp
Earl's Court basements, and a
few thousand people discoyering
sex and pot.
("London Evening Standard,"
12May 2000, P.13)
The choice of items signals that this is a
metonymic rather than a literal list.
Metonymy clusters exhibit the basic
metonymic principie of part-whole relations;
the examples are the parts, the more
general message is the whole. Our world
knowledge tells us there must have been
more to London in the 1960s than is
contained in the first three items. This is
confirmed by the next item being highly
specifíc, Australians in damp Earl's Court
basements; and finally, a few thousand
people discovering sex and pot, leaves the
reader in no doubt. The more prototypical
the examples, the more they signal that the
passage is metonymic and, generally, the
more powerful the effect.
The impact of figurative thought on
the larger scale of discourse gives rise to
phenomena which are different in form and
nature from those encountered at clause
leveI, with the result that metonyrny-in-
discourse phenomena are not always
immediately recognizable as examples of
what most people think of as metonymy.
This is not new. Jakobson makes
foundational statements about the role of
metonymy in communication in his classic
paper on aphasia, identifying two distinct
'peles' of communication, one metonymic,
the other metaphoric (Iakobson, 1956). In
the final section of the essay, he
characterizes prose, cinema and realism as
reflecting the metonymic 'way; and poetry,
theatre and surrealism the metaphoric way
(Jakobson, 1956, pp.76-79). Lodge takes
up Iakobsons distinction, referring instead
to metaphoric and metonyrnic 'modes' of
writing
(Lodge, 1977). He makes the important
observations that writing may be
metonymic at text levei but not necessarily
at surface leveI, and that 'metonymic
writing' does not necessarily contain
linguistic metonymies, consisting instead of
'literal' language and even metaphoric
expressions: "It is metonymic writing, not
metaphoric, even though it contains a few
metaphors and no metonymies; it is
metonymic in structure" (Lodge, 1977,
PP.98-99).
The independence of metonymic
language and metonymic thinking in
talk/text has also been noted by scholars in
more recent times. Gibbs distinguishes
between 'processing metonymic language'
and the 'metonymic processing of language'
(Gibbs, 1999, p.69), that is, between
recognizing individual expressions as
metonymies versus recognizing part-whole
thinking at discourse level; and points out
that comprehending individual expressions
which contain 'conventional metonymic
language' (what most people consider
metonymy to be) does not necessarily draw
on 'metonymic mappings' (Gibbs, 1999,
P.74). "The proper
study of metonymy', he writes, "surely
extends beyond looking at metonymic
language
alone", and that we need to look beyond
"metonyrny as a lexical phenomenon [... ]
to discover the ways that patterns of
metonymy in language reflect patterns of
metonymic thought" (Gibbs, 1999, p. 74).
For Gibbs, "speaking and understanding
indirect speech acts involves a kind of
metonymic reasoning, where people infer
wholes (a series of actions) from a part"
(Gibbs, 1994, P.352). Pragmatic
inferencing has been explored from this
perspective by a number of authors (e.g.,
Panther & Thornburg, 2003).
Just as procedures developed for
metaphor identification are not well
equipped for recognizing 'large scale'
phenomena such as extended metaphor,
so procedures for metonymy identification
are not well suited for recognizing
largescale metonymy phenomena, such as
'clusters' (in the sense that I am using the
term), as both require analysis at a macro-
level. Biernacka demonstrates that the
principle of metonymy is involved not just
in processing lexis but also on a larger
scale (Biernacka, 2013,p.208). She points
out that the system she has developed for
the identification of metonymy, operating in
a similar way to MlP and MlV by looking for
differences between the contextual and
basic meaning of lexical items/ phrases,
does not pick up metonymic thinking on a
larger scale, and identifies two phenomena
on this larger scale, 'metonymic shifting of
pronominal reference' and the 'metonymic
processing of scenarios and stories'.
Biernacka identifies a section of focus-
group data where there is intense activity
at the macro-level, which she calls a
'super-cluster, where five metaphor
clusters co-occur with a high number of
word-level metonymies (Biernacka,
2013,P.153).This is also the point where
the most controversial and emotional
topics are being discussed.

***

3.2 Metoymy chains

The idea of a 'chain' of metonymies has two

senses in the literature: a 'horizontal'
discourse sense and a 'vertical' virtual
sense. The vertical sense refers to multiple
metonymic mappings initiated by a single
lexical item, an inferred chain of concepts,
with each concept providing the vehicle for
the next. Various terms have been used to
describe this: 'chain of rnetonymies, e.g.,
idea-word-page-booklibrary (Reddy, 1993,
pp.186-187); '(inclusive) metonymic chain,
e.g., head-brain-thinking-mind-intelligence
(Dirven, 2002, PP.98, 103); 'chained
metonymies' (Hilpert, 2010); and
'metonymic chaíníng, e.g., glasses-
goalkeeper-Preston North End football
team (Littlemore, 2015,P.131).I am
concerned instead with the horizontal sense
of a chain, a 'linear' sequence of
metonymicallyrelated lexical items, linking
across a text and serving a discourse
function, to which Brdar-Szabó & Brdar give
the name 'textual metonymic chain' (2011,p.
229).
Viewing cohesion in terms of
metonymic relations has been discussed by
Stirling (1996), Al-Sharafi (2004), Brdar-
Szabó & Brdar (2011),Denroche (2015),
Littlemore (2015) and others. There are
many ways in which meaning relations
between lexical items set up metonymy
chains and networks in text, like through
meronymy, superordinancy, hyponymy,
antonymy, but it is synonymy which I want
to use to illustrate metonymy chains in this
section. ln the theoretical framework of this
article, meaning relations between
synonyms are seen as metonymic because
they involve parts and wholes; synonyms
are related to each other metonymically
because they share many meaning
components, and because recogmsmg
relations between synonyms involves
metonymic thinking. The extract below from
a self-help book explores the relationship
between Andrew andGwen:
Andrew handled his sensitivity and
reactivity somewhat differently.
Andrews
style was to turn a deaf ear to
Gwen. She referred to this as the
deep freeze. He
was civil, even polite, but
completely unavailable. Gwen had
learned it was best to
leave Andrew alone until he was
ready to interact. Trying to talk with
him when
he pulled back was like cornering a
fox, which will bite when trapped. It
was hard
for Gwen when Andrew walled her
out.
 (D. Schnarch, "Resurrecting sex:
Resolving sexual problems and
rejuvenating
your relatíonship," 2002, p. 42)
Each sentence in this passage enriches the
message with a new term to describe
emotional distance/isolation. A fuI! picture
of what being in a relationship with Andrew
felt like for Gwen is built up through a chain
of synonymous expressions (underlined).
The terms are not exact equivalents but
metonymical!y related, overlapping
sufficiently for the reader to process the m
as related.
In Halliday & Hasans account of
cohesion in English, a chain of synonyms is
one way 'reiteration' achieves 'lexical
cohesion' in text (Halliday & Hasan, 1976).
Using the term 'metonyrny chain', rather
than adopting Halliday & Hasan's
terminology (lexical cohesion/reiteration),
emphasizes that the function of this sort of
chain is not simply to re-refer, as Halliday &
Hasan suggest, but to enrich meaning
progressively as the text unfolds. The items
in the chain in the 'Andrew and Gwen' text
above, turn a deaf ear, unavailable, (not)
ready to interact and pull back, have
different associations, many of them
metaphoric, and do not merely represent
repetitions.
Metonymy, by its very nature, lends
itself to the realization of the progressive
enrichment of meaning. Kress maintains
that representation is always 'partial, partial
"in relation to the object or phenomenon
represented', and 'full' "in rei ation to the
sign-maker's interest at the moment of
making the sign" (Kress, 2010, P.71). Seto
divides metonymies into those which
involve specific-general or 'kind of' relations
(C type) and those which involve part-whole
or 'part of' relations (E type) (Seto, 1999),
while for Ruiz de Mendoza Ibáfiez & Diez
Velasco metonymies are oftwo types,
SOURCE-IN-TARGET and TARGET-IN-
SOURCE, facilitating, respectively, 'dornain
expansion' and 'domain reduction' (Ruiz de
Mendoza Ibáüez & Diez Velasco, 2002).
Thus, at every point in discourse, metonymy
makes available small shifts in meaning at
a very basic leveI. The possibilities are
further increased when discourse is
multimodal, each 'mode' offering new
potential for shifting meaning and
emphasizing different aspects of a domain,
such as in pictorial material of comics and
cartoons (Forceville, 2008, P.475),
'multimodal metonymy' in advertising
billboards and feature films, where visual
metonymies are 'source-intarget' rather
than 'target-in-source' (Forceville, 2009),
and 'metonymic chains' and 'double
metonymies' expressed multimodally in lCT
advertisements (Hidalgo & Kraljevic, 2011).
In a corpus of printed advertisements,
Pérez-Sobrino found source and target
domains were cued visually, verbally and
verbopictorially, and that metaphtonymy
was the most frequently used 'conceptual
operatíon, metonymy offering a point of
access to a domain and metaphor providing
connotational mappings (Pérez-Sobrino,
2016).
Al-Sharafi's cognitive-semiotic
approach puts metonymy at the centre of
communication and characterizes 'the sign'
itself as metonymic. He considers ali
Halliday & Hasans categories of cohesion,
grammatical as well as lexical, to be
metonymic, giving 'texture' through 'surface
text ties, while also creating 'deeper'
cognitive metonymic links: "I do not discuss
cohesion as a set of surface text ties only,
but from the point of view of its creation by
metonymic relations in text" (Al-Sharafi,
2004, p.110). Al-Sharafi feels that
"metonymy accounts for the relations of
lexical cohesion in a more satisfactory way
than the term 'Iexical cohesion' itself" (Al-
Sharafi, 2004, p.126). Stirling examines
cohesion in terms of metonymy but
concentrates on grammatical rather than
lexical relations (Stirling, 1996). The term
'metonymic anaphora' in her work refers to
contexts where a pronoun triggers an
aspect of a lexical item different from the
one initially intended. She illustrates this
with an example from a text about Weight
Watchers (an organization which organizes
weight-Ioss programmes), in which the
lexical item Weight Watchers has the sense
of 'institutíon, but late r in the text the
pronoun they triggers the sense of 'people'
(Stírling, 1996, p.69). Stirling maintains that
inanimate to animate shifts such as these
appear to be unproblematic in terms of
processing in the studies she reviews
(Stirling, 1996, P.71).
Brdar-Szabó & Brdar see the
importance of metonymy in providing
cohesion across text in their discussion of
'metonymic chains' (Brdar-Szabó & Brdar,
2011). Such chains not only enhance
coherence and cohesion but also allow
"plenty of conceptual maneuvering room"
(Brdar-Szabó & Brdar, 2011, PP.245-246).
Brdar-Szabó & Brdar distinguish between
'textual' and 'conceptual' metonymic chains.
In a 'textual metonymic chain' the same
lexeme is repeated across a text and
different aspects of the lexical item are
highlighted each time, allowing 'shifts'
"between subdomains within a single
domain matrix, picking different target
meanings at different points in a text, while
using a single lexeme as a metonymic
source" (Brdar-Szabó & Brdar, 2011, pp.
238-239). Ancient Rome can trigger various
meanings - the territory of the Roman
Empire, the city of Rome in Roman times,
the influence, customs and culture of the
Ancient Romans - depending on where it
appears in a text (author's example). A
'conceptual' metonymic chain in contrast
(similar to my sense of'metonymy chain'),
consists of different lexical items which
15
develop a single mental concept as the
reader progresses through the text, a series
of different metonymic sources "unified by
common metonymic targets" (Brdar-Szabó
& Brdar, 2011, P.232).
Biernacka presents data from focus
groups on terrorism to show shifts in
meaning of the pronouns they, we and you
across text, for which she coins the term
'metonymic shifting of pronominal
reference, maintaining that a discourse-
dynamic approach is needed to reveal the
"cornplex, dynarnic, contextand process-
dependent nature" of metonymy
(Biernacka, 2013, P.231). Kimmel gives an
example of a text in which such a chain is
set up within the source domain of an
extended metaphor, the lexical items volte
face, U-turn, withdrawing forming a chain of
metonymically-related iterns, though
Kimmel describes this not as a metonymy
chain but as "cohesion relations between
metaphors"
(Kimmel, 2012, P.34).

***

3.3 Extended metonymy

The third phenomenon I consider in this

overview of metonymy in discourse is
'extended metonyrny' I am using the term
here to refer to instances where a number
of novel linguistic metonyrnies, ali deriving
from the same conceptual metonymy, occur
together in dose proximity. Gibbs calls
these 'contextual expressions' and gives an
example in which the desirability of future
roommates is discussed by drawing
creatively on the metonymy POSSESSION
FOR PERSON, the individuaIs being
referred to via their possessions, steam
iron, stereo, electric typewriter, etc. (Gibbs,
1994, P.334). If the famous Ham sandwich
wants his check example were extended
within a text to other people in the
restaurant, this would be an example of
extending the FOOD ORDER FOR
PERSON metonymy. Similarly, referring to
various people in a hospital ward by the
conditions they are suffering would involve
an extension of CONDITIO FOR PERSON;
or if the injuries that players suffer during a
football season, such as knee, neck and
groin, were used to
identify the players via the conceptual
metonymy INJURY FOR PLAYER. The
lexical
items are novel in the sense that the
meanings they have in these contexts
would not be reported in any dictionary.
Dancygier & Sweetser, under the
heading 'Extended metonymy and
viewpoint,
consider an example from "Passage to
[uneau" by Ionathan Raban in which books
on a sailing boat are thrown off the shelves
onto the floor during rough seas (Dancygier
& Sweetser, 2014, PP.194-19S). A number
of metonymies are involved, BOOK-TITLE
FOR THE PHYSICAL BOOK, AUTHOR'S
NAME FOR THE PHYSICAL BOOK,
BOOK-TITLE FOR IDEAS IN THE BOOK
and AUTHOR'S NAME FOR THE AUTHOR
AS A PERSON, which then give rise to
metaphoric language such as unlikely
tangle, pages gaping, jackets half-off and
chance couplings
Extended metonymy involves novel
rather than conventional expressions,
deriving from the same conceptual
metonymy and occurring together in the
same section of text; this parallels
'extended metaphor' (Section 2.3), where a
number of different linguistic expressions
derive from the same conceptual metaphor.
There is an important difference, however,
as extended metaphor is more easily
detected using identification procedures,
such as those discussed above, than would
be the case for extended metonymy. The
reason for this is that conceptual
metaphors, such as GOOD 15 UP or LIFE
IS A JOURNEY, pattern lexis according to
specific domains; while the patterns
organized by conceptual metonymies, such
as OBJECT FOR PERSON or INANIMATE
FOR ANIMATE, indicate far more
generalized lexical domains. For this
reason, metonymy lacks what Handl calls
the "creative potential" of metaphor (Handl,
2011, pp. 89-90), novel linguistic
metonymies being the result of novel
contexts rather than the exploitation of
conventional mappings in novel ways, as is
the case for metaphor. Basic levei
metaphors, such as
GOOD IS UP, are closer to image schemas
and the direct embodiment of our sensory
experience of the physical world, and often
have metonymic origins; but, although
'basic, they pattern lexis in ways which are
more predictable than is the case for
metonymy. Conceptual metonymies also
range from models which are more 'basic'
(and closer to image schemas) to those
which are less basic, PART FOR WHOLE,
for example, being more primary than
INJURY FOR PLAYER. The metonymies
behind metonymy clusters and metonymy
chains are more basic, essentially PART-
WHOLE, than those giving rise to extended
metonymy.
ln this section, I have indicated that
metonymic thinking, like metaphoric
thinking, frequently plays a significant role
not only at, and below, the levei of the
clause, but also in organizing language at
the levei of the whole text. Not only does
metonymy play a powerful role in
discourse, but it has many different
manifestations and functions. It is well
established that the relationship between
cognitive aspects of metonymy and
linguistic manifestations of metonymy are
complex and operate at different levels,
with the consequence that figurative
thought does not always manifest itself as
figurative language. Having considered
metaphor in discourse and metonymy in
discourse as independent phenomena in
this and the previous section, I now go on
to look at the interaction of these
phenomena in talk/text.

***

4. Text metaphtonymy

Goossens identifies four types of

metaphtonymy, four ways in which
metonymy and metaphor 'combine' and
'intertwine' at clause levei (Goossens,
1990). These are paired into: integrated
metaphtonymy, which comprises
'metonymy within metaphor' and 'metaphor
within metonymy'; and cumulative
metaphtonymy, which comprises 'metaphor
from metonymy' and 'metonymy from
metaphor' (Goossens, 1990, p. 338). I will
consider only integrated metaphtonymy as
it is here that metonymy and metaphor
combine but remain distinct. The word
'within' in 'metonymy within metaphor' and
'metaphor within metonymy' is key, as it
pinpoints the salient notion that both
metonymy and metaphor are present, but
that
there is a scalar difference between the
two elements, that two levels of magnitude
are involved. In cumulative metaphtonymy,
'from' indicates a process of derivation
where either metonymy or metaphor is the
'end product' or 'result' (Goossens, 1990,
P.338). While Goossens is concerned with
strings of words of c1ause length or
shorter, I am looking at how metonymy is
embedded in metaphor and metaphor is
embedded in metonymy on the larger
scale of the whole text. For this I am using
the term 'text metaphtonymy; while
retaining Goossens' descriptors 'metonymy
within metaphor' and 'metaphor within
metonymy' in the discussion below.

***

4.1 Metonymy within metaphor

To illustrate 'metonymy within metaphor' at

text level, I revisit the Olympics Games text
discussed in Section 2.3. The term
'metonymy within metaphor' indicates
metaphor organizing a larger unit within
which metonymy is present as a smaller
unit, or, as Goossens puts it, "a
metonymically used entity is embedded in a
(complex) metaphorical expression"
(Goossens, 1990, P.336). In the Olympics
Games text, I am taking the whole text to be
the larger unit, organized by the extended
metaphor LONDON IS A FLAT, and the
smaller unit the 'metonymy c1uster'
embedded within it, the tasks which have to
be carried out before the games begin: We
have litter to pick, graffiti to scrub, and
flowers to planto The Olympics Games text
provides an example of a 'metonymy
cluster' within an 'extended metaphor, and
therefore a 'metonymy within metaphor'
type of text
metaphtonymy.
The metonymy c1uster in this
example consists of three items - litter to
pick,
graffiti to scrub,flowers to plant - which
express the target domain ofthe extended
metaphor, LONDON; but a metonymy
c1uster could equally well draw from the
source dornain, as is the case in the spoken
text below, part of an IT class, where the
extended metaphor TEXT MANIPULATION
IS PAINTING is organizing the text at the
whole-text level.
My 'l-beam' is carrying a paint brush,
so when I click on the mouse I know I
will
reformat the highlighted texto There is
no point putting the paint brush in the
paint and then putting it back in the
pot. You want to paint something, a
fence, a
door, a waIl ar something. (IT training
session at a London University,
adapted)

The instructor uses the 'metonymy

cluster' You want to paint something, a
fence, a doar, a wall ar something. The
students are told that words in their
doeuments need to be highlighted for the
text-formatter tool to work, but the
exampies in the cluster, a fence, a doar, a
wall ar something, are from the source
domain, PAINTING, rather than the target
domain of TEXT MANIPULATION. The
Olympic Games and IT Instruction texts
thus present a further distinction within text
metaphtonymy, representing two types
of'metonymy cluster within extended
metaphor, one in which the cluster is set
up by the target domain, the other by the
source domain. A further discourse
pattern, 'metonymy within metonymy within
metaphor, which involves a second levei of
metonymy, is iIIustrated by one of the
'Welcome to Holland!' texts discussed by
Semino et aI. (Semino et aI., 2013, P.53).
Here, a metonymy cluster, Coliseum,
Sistine Chapel, gondolas, is employed to
represent 'Italy, which in turn stands for the
larger category of 'ali first-choice travei
destinations', which in turn cues the source
domain TRAVEL of the extended metaphor
PARENTING IS TRAVEL.

***

4.2 Metaphor within metonymy

1 now iIIustrate the second of the two

types of integrated metaphtonymy,
'metaphor within metonymy, using two
examples, one in which metonymy is
present as 'metonymy clusters' and the
other as 'metonymy chains' To iIIustrate
the former, 1use an example from a poem
by the English poet Philip Larkin, 'Toads
Revisited' (Larkin, P., 1964, "The Whitsun
Weddíngs," pp.18-19). In this poem,
metonymic clusters are used to evoke a
number of different contexts: 'the park, 'the
people you find in the park', 'what those
people do during the day' and 'the offíce'
Looking closer, we find there are local
metaphors occurring within the larger
frame of a metonymy cluster, a
phenomenon noted by Lodge, who
maintains
that most metonymic texts "contain a good
deal of local metaphor" (Lodge, 1977,
p.rn). The people in the park include
'clerks' and 'outpatients': hare-eyed clerks
with the jitters and wax-fleshed outpatients
still vague from accidents. Hare to describe
'eyes' and wax to describe 'flesh' are words
used metaphorically; thus, we have an
example of the 'metaphor within metonymy
(cluster)' type of text metaphtonymy.
A further levei of complexity
becomes apparent when we look at the
poem in its entirety. On and above the
metonymy clusters there is a further
metaphoric layer, the WORK IS A TOAD
metaphor which organizes the poem as a
whole. This gives a hierarchical structure
with three layers, the 'metonymy clusters' in
the middle serving both as smaller units
within the overall metaphoric framework of
the poem and larger units in which local
metaphors are embedded. The two types of
integrated metaphtonymy, 'metaphor within
metonymy' and 'metonymy within
rnetaphor, are found one within the other.
Lodge recognizes this tripledecker,
metaphor-metonymy-metaphor structure in
other Larkin poems, "The Whitsun
Weddings" and "Church Going," where
local metaphors are embedded in
metonymic writing and the overall
framework of the poem is metaphoric
(Lodge, 1977, pp. 217-218). This we might
designate 'metaphor within metonymy
within metaphor'
 To illustrate metaphor within
metonymy where metaphor occurs within a
metonymy chain, I return to the 'Andrew'
text (Section 3.2). In this text, a 'metonymy
chain' is set up through a string of
synonyms which runs through the extract,
establishing cohesion as well as adding to
meaning item by item. As some of the
items in the chain are metaphoric, i.e., to
turn a deaf ear, the deep freeze, pulled
back, walled her out, we have an example
of local metaphor occurring within a
metonymy chain. In Goossens' dictionary
data, metaphor within metonymy is
"extremely rare';' accounting for only one
example, while metonymy
within metaphor is "quite current"
(Goossens, 1990, P.336). Goossens
suggests the reason for this asymmetry is
the tendency for metaphor to 'metaphorize'
the expression in which it is found:
''Ametaphor inserted into a metonym would
seem to metaphorize the whole, whereas a
metonym integrated into a metaphor does
not appear to have the power to
metonymize the rnetaphor" (Goossens,
1990, p. 338). At discourse levei, however,
a different picture emerges: on the larger
scale of the whole text, not only does
metaphor commonly occur within
metonymic writing, but there does not
seem to be a tendency for the metaphoric
elements to metaphorize the whole.

***

5. Conclusion

One of the lessons learned from the 'cognitive turn' is that metonymy and metaphor are not
just text phenomena but primarily about how we think; and that, if metonymy and metaphor
are fundamentally about thought, they can potentially have an impact on any size of unit of
language, from the very small to the very large, from short word-strings to long stretches of
language, and can aiso be expressed multimodally. In this article, I have offered a framework
for overviewing the various ways figurative thought manifests itself in speech and writing by
looking at metonymy in discourse in terms of clusters, chains and extended metonymy and
metaphor in discourse in terms of clusters, chains and

1. Metaphor within metonymy is mistakenly given as metaphor from metonymy in the abstract ofGoossens'
(1990) article, but appears corrected in the 2003 reprint (Goossens, 2003).

extended metaphor.' I then used this framework to demonstrate how metonymic reasoning
and metaphoric reasoning combine in text metaphtonymy and the many forms it can take.
Hierarchal metaphor-metonymy-metaphor organizations in text have also been discussed.
 The different types of text metaphtonymy discussed above involve only three of the
metaphor and metonymy in discourse phenomena described in Sections 2 and 3, namely
'extended metaphor, 'metonymy clusters' and 'metonymy chains' I suggest that many more
metonymy-metaphor combinations are possible, though certain combinations offer greater
opportunities for text metaphtonymy. I have shown that extended metaphor, metonymy
clusters and metonymy chains have the capacity for setting up larger-scale structures within
text and interactions at discourse level, but there is no reason in principle why the remaining
three phenomena, 'metaphor clusters, 'metaphor chains' and 'extended metonyrny; could
not also form text metaphtonymies, though, as they operate on a smaller scale, the
interactions will tend to be more along the lines of Goossens' clause-level
examples.
I hope the contribution made by the present study may suggest the direction in which
further research in this field might take and ways in which these ideas might be applied.
Short, and often self-contained, examples have been given in this article for clarity of
explanation, but the phenomena discussed are to be found operating in longer texts, across
whole books and between texts. Text metaphtonymy is undoubtedly interesting in its own
right as a meaning-making phenomenon, and the motivation for the present article has been
to investigate it as such, but a further motivation for studying text metaphtonymy is to explore
the implications it has for training language professionals. What the experienced practitioner
does automatically, the novice needs to learn. Those training to be journalists, speech
writers, copywriters, text editors, language teachers, translators and interpreters, among
others, would all benefit, I feel, from the explicit teaching of the figurative text-phenomena
discussed in this article. Further research may then embrace more extensive studies which
are both systematic and domain specific.

2. Elsewhere (Denroche, 2015), I have given four of these phenomena other names, as my purpose there
was different: to contrast the use of metonymy and metaphor in changing register with their use in patterning
lexis. The terms correspond as follows (present article first, then the 2015 publication): metonymy cluster =
discourse metonymy, metaphor cluster = discourse metaphor, metonymy chain = textual rnetonymy,
extended metaphor = textual metaphor.

References
Al-Sharafi, A. (2004). Textual metonymy: A semiotic approach. Basingstoke: Palgrave.
https:/ 1doLorg/10.1057/9781403938909
Barcelona, A. (Ed.). (2000). Metaphor and metonymy at the crossroads. Berlin: Mouton de
Gruyter.
Benczes, R., Barcelona, A., & Ruiz de Mendoza, F. (Eds.). (2011). Defining metonymy in
cognitive linguistics: Towards a consensus view. Amsterdam: [ohn Benjamins.
httpsv/doi.orq/to.tozs/hcp.za
Biernacka, E. (2013). A discourse dynamics investigation of metonymy in talk.
Unpublished PhD
thesis. Milton Keyes: The Open University.
Brdar-Szabó, R., & Brdar, M. (2011). What do metonymic chains reveal about the nature
of metonymy? In R. Benczes, A. Barcelona & F. Ruiz de Mendoza (Eds.), Defining
metonymy
in cognitive linguistics: Towards a consensus view (pp. 217-248). Amsterdam: [ohn
Benjamins. httpsv/doi.orq/to.rozs/hcp.za.tabrd
Cameron, L. (1999). Identifying and describing metaphor in spoken discourse data. In
L. Cameron & G. Low (Eds.), Researching and applying metaphor (pp. 105-132).
Carnbridge, UK: Cambridge University Press.
https://doi.org/1O.1017/CB09781139524704.009
Cameron, L. (2003). Metaphor in educational discourse. London: Continuum.
Cameron, L. (2008). Metaphor and talk. In R. Gibbs (Ed.), The Cambridge handbook of
metaphor and thought (pp. 197-211). Cambridge, UK: Cambridge University Press.
https:lldoi.org/10.1017/CB09780511816802.013
Cameron, L. (2010). The discourse dynamics framework for metaphor. TnL. Cameron &
R. Maslen (Eds.), Metaphor analysis: Research practice in applied linguistics, social
sciences and humanities (pp. 77-94). London: Equinox.
Cameron, 1., & Deignan, A. (2006). The emergence of metaphor in discourse. Applied
Linguistics, 27(4), 671-690. httpsv/doi.orq/to.tosj/applin/arnlojz
Cameron, 1., & Low, G. (2004). Figurative variation in episodes of educational talk and
text.
European [ournal of English Studies, 78(3), 355-377-
https://doi.org/10.1080/1382557042000277430
Cameron, 1., Low, G., & Maslen, R. (2010). Finding systematicity in metaphor use. In
L. Cameron & R. Maslen (Eds.), Metaphor analysis: Research practice in applied
linguistics,
social sciences and humanities (pp. 116-146). London: Equinox.
Cameron, 1., & Maslen, R. (Eds.). (2010). Metaphor analysis: Research practice in applied
linguistics, social sciences and humanities. London: Equinox.
Cameron, 1., Maslen, R., Todd, Z., Maule, J., Stratton, P., & Stanley, N. (2009). The
discourse
dynamics approach to metaphor and metaphor-led analysis. Metaphor and Symbol, 24(2),
63-89. https://doLorg/10.1080/10926480902830821
Cameron, 1., & Stelma, J. (2004). Metaphor cJusters in discourse. lournal of Applied
Linguistics,
1(2),107-136. https://doLorg/10.1558/japI.2004.1.2.107
Dancygier, B., & Sweetser, E. (2014). Figurative language. Carnbridge, UK: Cambridge
University Press.
Darian, S. (2000). The role of figurative language in introductory science texts.
International
[ournal of Applied Linguistics, 10(2), 163-186.
https://doLorg/10.1111/j.1473-4192.2000.tbo0147.X
22 Charles Denroche
Deignan, A. (2005a). A corpus linguistic perspective on the relationship between
metonymy
and metaphor. Style, 39(1), 72-91.
Deignan, A. (2005b). Metaphor and corpus linguistics. Amsterdam: [ohn Benjamins.
httpsv/doi.orq/to.tozs/celcr.õ
Deignan, A., Littlemore, J., & Semino, E. (2013). Figurative language, genre and register.
Cambridge, UK: Cambridge University Press.
Denroche, C. (2015). Metonymy and language: A new theory oflinguistic processing. New
York/London: Routiedge.
Dirven, R. (2002). Metonymy and metaphor: Different mental strategies of
conceptualisation.
ln R. Dirven & R. Põrings (Eds.), Metaphor and metonymy in comparison and contrast
(pp. 75-111). Berlin: Mouton de Gruyter. https://doLorgho.1515/9783110219197.75
Dirven, R., & Põrings, R. (Eds.). (2003). Metaphor and metonymy in comparison and
contrasto
Berlin: Mouton de Gruyter. https://doLorg/10.1515/9783110219197
Forceville, C. (2008). Metaphor in pictures and multimodal representations. ln R.W. Gibbs
(Ed.), The Cambridge handbook of metaphor and thought (pp. 462-482). Cambridge, UK:
Cambridge University Press. https://doLorgho.1017/CB09780511816802.028
Forceville, C. (2009). Metonymy in visual and audiovisual discourse. ln E. Ventola &
A. Guijarro (Eds.), The world told and the world shown: Issues in multisemiotics (pp.
56-74). Basingstoke: Palgrave MacMillan.
Gibbs, R. (1994). The poetics of mind: Figurative thought, language and understanding.
New
York: Cambridge University Press.
Gibbs, R. (1999). Speaking and thinking with metonymy. ln K. Panther & G. Radden
(Eds.),
Metonymy in language and thought (pp. 61-76). Amsterdam: [ohn Benjamins.
httpsv/dol.orq/ro.rozs/hcp.a.oaqlb
Goatly, A. (1997). The language of metaphors. London: Routiedge.
Goossens, L. (1990). Metaphtonymy: The interaction ofmetaphor and metonymy in
expressions for linguistic action. Cognitive Linguistics, 1(3), 323-340.
https://doLorgho.1515/CogI.1990.1.3.323
Goossens, L. (2003). Metaphtonymy: The interaction of metaphor and metonymy in
expressions for linguistic action. ln R. Dirven & R. Põrings (Eds.), Metaphor and
metonymy in comparison and contrast (pp. 349-377). Berlin: Mouton de Gruyter.
Halliday, M., & Hasan, R. (1976). Cohesion in English. London: Longman.
Handl, S. (2011). Salience and the conventionality of metonymies. ln S. Handl & H.
Schmid
(Eds.), Windows to the mind: Metaphor, metonymy and conceptual blending. (pp. 85-114).
Berlin: Mouton de Gruyter.
Hidalgo, L., & Kraljevic, B. (2011). Multimodal metonymy and metaphor as complex
discourse
resources for creativity in lCT advertising discourse. ln F. Gonzálvez-García, S. Pena, &
L. Pérez (Eds.), Metaphor and metonymy revisited beyond the contemporary theory of
metaphor. Special issue ofthe Review ofCognitive Linguistics, 9(1), 153-178.
Hilpert, M. (2010). Chained metonymies. ln J. Newman & S. Rice (Eds.), Experimental and
empirical methods in cognitive functional research (pp. 181-194). Stanford, CA: CSLI.
[akobson, R. (1956). Two aspects oflanguage and two types of aphasic disturbances. ln
R. [akobson & M. Halle, Fundamentais oflanguage (pp. 53-82). The Hague: Mouton de
Gruyter.
Kimmel, M. (2010). Why we mix metaphors (and mix them well): Discourse coherence,
conceptual metaphor, and beyond. [ournal of Pragmatics, 42, 97-115.
https://doi.orgho.1016/j.pragma.2009.05.017
Text metaphtonymy 23
Kimmel, M. (2012). Optimizing the analysis of metaphor in discourse. Review of Cognitive
Linguistics, 10(1), 1-48. httpsv/dol.orq/ro.tozs/rcl.to.t.otklm
Koller, V. (2003). Metaphor c1usters, metaphor chains: AnaJyzing the multifunctionality of
metaphor in texto metaphorik. de, S, llS-134.
Kõvecses, Z. (2002). Metaphor: A practical introduction. Oxford: Oxford University Press.
Kress, G. (2010). Multimoda/ity: A social semiotic approach to contemporary
communication.
London: Routledge.
Lakoff, G., & Iohnson, M. (1980). Metaphors we /ive by. Chicago, IL: University ofChicago
Press.
Littlemore, J. (201S). Metonymy: Hidden shortcuts in language, thought and
communication.
Cambridge, UK: Cambridge University Press. https://doi.orgI10.1017/CB09781107338814
Lodge, D. (1977). The modes of modern writing: Metaphor, metonymy and the typology of
modern /iterature. London: Arnold.
Low, G. (2008). Metaphor and positioning in academic book reviews. In M. Zanotto,
L. Cameron & M. Cavalcanti (Eds.), Confronting metaphor in use: An applied /inguistic
approach (pp. 79-100). Amsterdam: Iohn Benjamins.
https://doi.orgI10.107s/pbns.173.0610w
Musolff, A. (2000). PoliticaJ imagery ofEurope: A house without exit doors. [ournal of
Multilingual and Multicultural Development, 21(3). 216-229.
htt ps://doi.orq/i 0.1080/01434630008666402
Panther, K., & Radden, G. (Eds.). (1999). Metonymy in language and thought. Amsterdam:
Iohn
Benjamins. httpsv/dol.orq/ro.tozs/hcp.a
Panther, K., & Thornburg, L. (Eds.). (2003). Metonymy and pragmatic inferencing.
Amsterdam:
[ohn Benjamins. httpsv/doi.orç/to.tozs/pbns.rn
Panther, K., Thornburg, L., & Barcelona, A. (Eds.). (2009). Metonymy and metaphor in
grammar. Amsterdam: Iohn Benjamins. httpsi//doi.orç/to.rozg/hcp.zs
Pérez-Sobrino, P. (2016). Multimodal metaphor and metonymy in advertising: A corpus-
based
account. Metaphor and Symbol; 31(2), 73-90.
Pragglejaz Group. (2007). MIP: A method for identifying metaphoricaJly used words in
discourse. Metaphor and Symbol, 22(1), 1-39.
https://doi.orgI10.10801109264807093367S2
Reddy, M. (1993). The conduit metaphor: A case of frame conflict in our language about
language. In A. Ortony (Ed.), Metaphor and thought (znd ed.) (pp. 164-201). Cambridge,
UK: Cambridge University Press. https://doi.orgI10.1017/CB0978113917386s.012
Ruiz de Mendoza Ibánez, F.,& Diez Velasco, O. (2002). Patterns of conceptuaJ
interaction. In
R. Dirven & R. Põrings (Eds.), Metaphor and metonymy in comparison and contrast (pp.
489-S32). Berlin: Mouton de Gruyter. https://doi.orgI10.1S1S/9783110219197A89
Semino, E. (2008). Metaphor in discourse. Cambridge, UK: Cambridge University Press.
Semino, E., Deignan, A., & Littlemore, J. (2013). Metaphor, geme, and
recontextuaJization.
Metaphor and Symbol, 28(1), 41-S9. https://doi.orgI10.1080110926488.2013.742842
Seto, K. (1999). Distinguishing metonymy from synecdoche. In K. Panther & G. Radden
(Eds.),
Metonymy in language and thought (pp. 91-120). Amsterdam: Iohn Benjamins.
httpsv/dot.orç/io.rozs/hcp.a.oeset
Steen, G. (2002). Identifying metaphor in language: A cognitive approach. Style. 36(3)
386-407.
Steen, G. (2007). Finding metaphor in grammar and usage. Amsterdam: Iohn Benjamins.
https.z/dot.orq/io.iozs/cekr.ro
Steen, G. (2014). Why figurative thought and language are not enough: On the crucial role
of
metaphor in communication. Paper given at ist lnternational Symposium on Figurative
Thought and Language. April2S-26, 2014, Aristotle University ofThessaJoniki.
24 Charles Denroche
Steen, G., Dorst, A., Berenike Herrmann, J., Kaal, A., Krennmayr, T., & Pasma, T. (2010).
A
method for linguistic metaphor identification: From MIP to MIPVU. Amsterdam: [ohn
Benjamins. httpsv/doi.orq/to.tozs/celcr.ta
Stefanowitsch, A., & Gries, S. (2006). Corpus-based approaches to metaphor and
metonymy.
Berlin: Mouton de Gruyter. httpsJ/doi.orgI10.1S1S/978311019989S
Stirling, L. (1996). Metonymy and anaphora. In W. de Mulder & L. Tasmowski (Eds.),
Coherence and anaphora (pp. 69-88). Amsterdam: [ohn Benjamins.
Address for correspondence
Charles Denroche
University ofWestminster
32-38 Wells Street
London W1T 3UW
UK
denrocc@westminster.ac.uk
Biographical notes
Charles Denroche is a Senior Lecturer with the Department of English, Linguistics and
Cultural Studies at the University ofWestminster, London. He studied at the universities
ofOxford, Florence, Düsseldorf, UCL-IoE and Westminster. He lectures and researches in
the areas of semiotics, metonymy and metaphor, semantics, discourse analysis and
translation/interpreting studies. He has worked as a language teacher, lexicographer and
translator in the UK and on the continent.
rly chains nor extensions